Index: llvm/trunk/include/llvm/IR/IntrinsicsNVVM.td
===================================================================
--- llvm/trunk/include/llvm/IR/IntrinsicsNVVM.td
+++ llvm/trunk/include/llvm/IR/IntrinsicsNVVM.td
@@ -729,20 +729,16 @@
                                       [IntrReadWriteArgMem, NoCapture<0>]>;
 
 // Bar.Sync
-//
-// TODO: Remove NoDuplicate here after fixing up LLVM to handle convergent
-// properly.  See discussion in http://reviews.llvm.org/D16941 and
-// http://reviews.llvm.org/D12246.
   def int_cuda_syncthreads : GCCBuiltin<"__syncthreads">,
-      Intrinsic<[], [], [IntrNoDuplicate, IntrConvergent]>;
+      Intrinsic<[], [], [IntrConvergent]>;
   def int_nvvm_barrier0 : GCCBuiltin<"__nvvm_bar0">,
-      Intrinsic<[], [], [IntrNoDuplicate, IntrConvergent]>;
+      Intrinsic<[], [], [IntrConvergent]>;
   def int_nvvm_barrier0_popc : GCCBuiltin<"__nvvm_bar0_popc">,
-      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrNoDuplicate, IntrConvergent]>;
+      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrConvergent]>;
   def int_nvvm_barrier0_and : GCCBuiltin<"__nvvm_bar0_and">,
-      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrNoDuplicate, IntrConvergent]>;
+      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrConvergent]>;
   def int_nvvm_barrier0_or : GCCBuiltin<"__nvvm_bar0_or">,
-      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrNoDuplicate, IntrConvergent]>;
+      Intrinsic<[llvm_i32_ty], [llvm_i32_ty], [IntrConvergent]>;
 
   // Membar
   def int_nvvm_membar_cta : GCCBuiltin<"__nvvm_membar_cta">,
Index: llvm/trunk/test/Feature/intrinsic-noduplicate.ll
===================================================================
--- llvm/trunk/test/Feature/intrinsic-noduplicate.ll
+++ llvm/trunk/test/Feature/intrinsic-noduplicate.ll
@@ -1,9 +1,9 @@
 ; RUN: llvm-as < %s | llvm-dis | FileCheck %s
 
-; Make sure LLVM knows about the convergent and noduplicate attributes on the
+; Make sure LLVM knows about the convergent attribute on the
 ; llvm.cuda.syncthreads intrinsic.
 
 declare void @llvm.cuda.syncthreads()
 
 ; CHECK: declare void @llvm.cuda.syncthreads() #[[ATTRNUM:[0-9]+]]
-; CHECK: attributes #[[ATTRNUM]] = { convergent noduplicate nounwind }
+; CHECK: attributes #[[ATTRNUM]] = { convergent nounwind }